One sample has quite different # of reads from different lanes.

Question

[jkim]

Hello,

This may not be related to bioinformatics, but I demultiplexed one illumina sequence run and noticed that one sample has quite different the # of reads from different lanes. (used the same index sequence in the samplesheet)

sample1 from lane1,3,4 : 0 reads

sample1 from lane2 : 8 million reads.

Would you mind explaining why it happened? It's nanostring sequence data. If you need more info about the libraries, I will definitely ask questions to my coworkers who prepared the libraries and ran the sequencer.

Answer 1

[GenoMax]

Which sequencer is this from? If the same pool did not run on multiple lanes then this would be the observation. Perhaps you received/created a wrong samplesheet. I suspect this to be case since there are 0 reads in other lanes.

Based on 4 lanes this is likely NovaSeq 6000, which can be run in XP mode where 4 separate pools can go on 4 lanes (or you could load the same pool 1 to 4 times).

[jkim]

Yes, you are correct. Novaseq instrument with S4 flowcell. I think I heard that XP mode was used. Yes, I will look into the samplesheet again.

So there could be possiblities that 4 seprate pools have something to do with the # of reads? (Sorry, I'm not familar with web lab work for the sequencers) even if I have a correct samplesheet?

Thank you for your insight!

[GenoMax]

If a single pool is loaded on all 4 lanes then one can see differing number of reads for any samples in that pool. But they will still be in the ballpark (within a few thousand) across the 4 lanes. No sample would have 0 reads in one or more lanes and 8 million in one lane.

If completely independent pools are loaded on 4 lanes then you would not expect the samples in a specific pool to show up on other lanes ... unless other lanes have samples that share the same index. Good bookkeeping is a MUST with sample pools since this information is being used to create the samplesheet.

[jkim]

Thanks a lot. It certainly makes sense!